agentleFS
MicrosoftKnown publishermicrosoft-foundry/foundry-agent-webapp/.github/skills/writing-bicep-templates/SKILL.md

writing-bicep-templates

Skill124 starsChanged 7 months agoMITView on GitHub

Provides Bicep coding standards for Azure infrastructure in this repository. Use when writing or modifying Bicep files, configuring Container Apps, setting up RBAC, or working with Azure resources.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Goal: Create consistent, secure Azure infrastructure

    line 8
  2. Very specific:

    Use resourceToken from uniqueString():

    line 12
  3. Very specific:

    Always add @description() and use @allowed() for constrained values:

    line 23
  4. Very specific:

    Use a user-assigned MI for ACR pull and OBO (avoids circular dependencies).

    line 46
  5. Very specific:

    Use guid() for names + specify principalType:

    line 61
  6. Distinctive:

    Use user-assigned MI for ACR pull (no admin credentials or secrets):

    line 98

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Bicep Coding Standards

Goal: Create consistent, secure Azure infrastructure

Naming Convention

Use resourceToken from uniqueString():

var token = toLower(uniqueString(subscription().id, environmentName, location))
name: '${abbrs.appContainerApps}web-${token}'  // ca-web-abc123

Exception: ACR requires alphanumeric only: cr${resourceToken}

Parameters

Always add @description() and use @allowed() for constrained values:

@description('Environment (dev, prod)')
param environmentName string

@description('Azure region')
@allowed(['eastus2', 'westus2'])
param location string = 'eastus2'

Outputs

Expose key identifiers for azd and other modules:

output containerAppName string = containerApp.name
output webEndpoint string = 'https://${containerApp.properties.configuration.ingress.fqdn}'
output identityPrincipalId string = containerApp.identity.principalId

Managed Identity

Use a user-assigned MI for ACR pull and OBO (avoids circular dependencies). Create it in the infrastructure module so its principalId is available before the Container App:

resource managedIdentity 'Microsoft.ManagedIdentity/userAssignedIdentities@2024-11-30' = {
  name: '${abbrs.managedIdentityUserAssignedIdentities}web-${resourceToken}'
  location: location
  properties: { isolationScope: 'Regional' }
}
output managedIdentityPrincipalId string = managedIdentity.properties.principalId

Attach to Container App with identity: { type: 'UserAssigned', userAssignedIdentities: { '${miId}': {} } }. Use MI for ACR pull via registries: [{ server: acr.loginServer, identity: miId }].

RBAC Assignments

Use guid() for names + specify principalType:

resource roleAssignment 'Microsoft.Authorization/roleAssignments@2022-04-01' = {
  name: guid(resource.id, principalId, roleId)
  properties: {
    roleDefinitionId: subscriptionResourceId('Microsoft.Authorization/roleDefinitions', roleId)
    principalId: principalId
    principalType: 'ServicePrincipal'
  }
}

Container Apps

Key settings: System identity + scale-to-zero + HTTPS only:

resource containerApp 'Microsoft.App/containerApps@2023-05-01' = {
  identity: { type: 'UserAssigned', userAssignedIdentities: { '${userAssignedIdentityId}': {} } }
  properties: {
    configuration: {
      ingress: {
        external: true
        targetPort: 8080
        allowInsecure: false
      }
    }
    template: {
      scale: { minReplicas: 0, maxReplicas: 3 }
    }
  }
}

ACR Pull Pattern

Use user-assigned MI for ACR pull (no admin credentials or secrets):

registries: [{
  server: containerRegistry.properties.loginServer
  identity: userAssignedIdentityId  // MI with AcrPull role
}]

Validation

az bicep build --file main.bicep
az deployment group what-if --template-file main.bicep

Project-Specific: Module Hierarchy

main.bicep (subscription scope)
├─ Resource group
├─ main-infrastructure.bicep (ACR + Container Apps Env + Log Analytics + User-Assigned MI)
├─ entra-app.bicep (SPA app + conditional OBO backend app with FIC + admin consent)
├─ main-app.bicep (Container App with MI-based ACR pull)
└─ RBAC (Cognitive Services User role via postprovision CLI)

Project-Specific: Container App Configuration

resource containerApp 'Microsoft.App/containerApps@2024-03-01' = {
  identity: {
    type: 'UserAssigned'
    userAssignedIdentities: { '${userAssignedIdentityId}': {} }
  }
  properties: {
    managedEnvironmentId: containerAppsEnvironmentId
    configuration: {
      ingress: {
        external: true
        targetPort: 8080
        allowInsecure: false
      }
      registries: [{
        server: containerRegistry.properties.loginServer
        identity: userAssignedIdentityId  // MI-based pull, no secrets
      }]
    }
    template: {
      containers: [{
        name: 'web'
        image: containerImage
        env: containerEnv  // Base env + conditional OBO env
        resources: { cpu: json('0.5'), memory: '1Gi' }
      }]
      scale: { minReplicas: 0, maxReplicas: 3 }
    }
  }
}

output fqdn string = containerApp.properties.configuration.ingress.fqdn
output identityPrincipalId string = containerApp.identity.principalId

Related Skills

  • deploying-to-azure - Deployment commands and hook workflow
  • writing-csharp-code - Backend configuration for Container Apps
  • troubleshooting-authentication - RBAC and managed identity debugging

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.